ConvertAxesToRealSpace: Algorithm to convert workspace Axes to detector units

[Nick Draper]

For the live viewing of data sometimes the instrument view is too slow and complex. Look at creating an algorithm to Convert the spectrum and TOF axis to various detector values. The TOF data will be integrated.

The resulting workspace will need to be freshly binned, probably using centre point rebinning to distribute the data.

[Nick Draper]

prime values for detector values are X, y, z, r, theta ,phi, detector2theta

[Nick Draper]

re #10066 ConvertAxesToRealSpace

Changeset: 8aec1ec6bebc6b8d1ac43b9fdc79fd1507882c0b

[Nick Draper]

re #10066 sort linux ambigous call

Changeset: 77f13d78f47878bf17d67be73073ae5ea1012751

[Nick Draper]

re #10066 sort linux warning

Changeset: 81bdc192ab9a4c77b0c36df347ec3d89dc5dc79f

[Nick Draper]

re #10066 usage full stop

Changeset: 5936fac2fcb30f8c4c95a1dbead5d57fbed8b0e7

[Nick Draper]

To Test

1. read alg documentation
2. load a workspace - LET, Merlin or any other instrument you understand the geometry of
3. Try this alg with various axes options

[Anders Markvardsen]

Doc, usage example, code, unit test added and nice pictures.

However, when I ran it on
isis\inst$\NDXSANS2D\Instrument\data\cycle_11_3\SANS2D00011109.nxs the log is flooded with warning. This is because sans2d contains a lot of workspace indices that points to nothing.

Hence I would suggest lowering a message like; "Could not find detector for workspace index 96768" to for example debug or information, or output nothing....

[Nick Draper]

re #10066 logging volume reduced

Changeset: 63d79dbf3b35b2fad84e69c82a9bf95913b1d102

[Andrei Savici]

TwoTheta and detector theta are the same in Mantid notation, so it is confusing to have both. Theta and phi have labels as radians, but what is on the axis is degrees (phi goes from -180 to 180)

[Nick Draper]

re #10066 fix units for theta and phi

Changeset: 5a580475d3cd3af5bccd96e6b8d1b69042c9684f

[Nick Draper]

I've corrected the units on theta and phi - good catch by Andre there.

I don't agree about theta and detector TwoTheta, they can be the same, but only if the instrument is set up with the sample at 0,0,0 and the beam coming down z. While this is the "norm" it is not the case throughout, and some instruments have chosen 0,0,0 for the moderator and y as the beam axis.

[Nick Draper]

re #10066 Documentation further improved

Changeset: 3d38f7e71de96308c78375feffceace322424c6d

[Karl Palmen]

If find the documentation is OK and the code review. It runs without crashing. I tried it on Merlin, but am not confident enough in my understanding of the geometry to be sure it is working as required, so I return to testing pool.

[Andrei Savici]

A few minor things for documentation:

• theta and phi are in degrees, while two theta and signed two theta are in radians. This should go in the rst file
• see the above point: so a new version of the ConvertAxesToRealSpace_Example.png image should be produced

[Nick Draper]

re #10066 documentation updates

Changeset: 8b71e8af76990058e20100ae89f44a1b73ee69b0

[Nick Draper]

documentation updates and new image done

[Andrei Savici]

Merge remote-tracking branch 'origin/feature/10066_ConvertAxesToRealSpace'

Full changeset: ​897157605340e939a837716b07b8474c8be8d905

[Stuart Campbell]

This ticket has been transferred to github issue ​10908